SOLVED Plex failure after major failure -- 21.08 Beta Fixed the issue

[exodus454]

I have managed to fix it a couple times I've seen it, I don't remember the specifics of what solved it. My recollection is it has something to do with docker since out of the box it uses docker networking.

Check the docker and docker.socket services for errors, make sure they're enabled and running, try stopping and starting them and see if that gets you anywhere.

Usually under normal circumstances you should see all the containers running under "docker ps" - try that too, sometimes there's clues

Also make sure you can "ping google.com", I've definitely had many issues with the default gateway just disappearing and dns without being instantly noticeable, doesn't hurt to check.

 

[Mentox]

For me the solution was: k3s kubectl taint nodes ix-truenas ix-svc-start:NoExecute-

This removes the taint on "ix-truenas" (the node name)

 

[truecharts]

For me the solution was: k3s kubectl taint nodes ix-truenas ix-svc-start:NoExecute-

This removes the taint on "ix-truenas" (the node name)

This is not a good move, that taint is added because the host has not been able to confirm the health of the service.
Also: You are aware you necro'ed a thread from months ago?

 

[Mentox]

This is not a good move, that taint is added because the host has not been able to confirm the health of the service.
Also: You are aware you necro'ed a thread from months ago?

It depends on what the root cause is obviously and if it is a single node your options are limited, removing a taint wont cause any extra "damage". If the underlying cause still prevents K3s from starting you will find out soon enough. In my case the issue was that the ix-applications/k3s folders didn't mount on time.

About "necro'ing" an old thread, should I create new topics or shall I rather not share a solution that might potentially help someone?

 

[truecharts]

It depends on what the root cause is obviously and if it is a single node your options are limited, removing a taint wont cause any extra "damage". If the underlying cause still prevents K3s from starting you will find out soon enough. In my case the issue was that the ix-applications/k3s folders didn't mount on time.

About "necro'ing" an old thread, should I create new topics or shall I rather not share a solution that might potentially help someone?

As you seem to be new to TrueNAS, it's important to note that these systems are designed so that only the middleware controls the underlaying software stack. Said taints are added (and removed!) by the middleware. Messing with it can (and quite often does) lead to unforseen consequences.

Often the beter choice is to force a GUI restart of the k3s subsystem via middleware, by changing the listen-interface in the Apps settings dialog, back and forth.

That should trigger 2 restarts/reloads of k3s via the middleware.

---

Generally speaking, commenting on BETA/RC threads from 6 months ago is not really appreciated, as these threads are kinda supposed to "die off" and not keep being revived by new solutions after half a year or more.